Artificial intelligence experts at the University of New England are helping the Commonwealth Department of Agriculture, Fisheries and Forestry prepare for any potential outbreak of foot-and-mouth disease (FMD) in Australia.
The UNE team, led by Professor A.S.M. Sajeev, is updating the Department’s software model of the spread of such an outbreak. “While the existing model, developed some years ago, would be useful at a regional level,” Professor Sajeev said, “the new model -developed initially at CSIRO and now here at UNE – is designed to provide a national perspective. Modelling at the national level can capture the implications or regional differences in climate, environment and livestock practices. This is very important for a country such as Australia.”
FMD has never reached Australia, but a recent outbreak in Japan has highlighted the need to be prepared. Using an advanced computing technique called “massive agent-based modelling”, the UNE team has been able to create a model that takes account of the interactions of individual animals (the “agents”) with each other and with the environment.
“The model is designed to facilitate decision making and response strategies in the event of a disease outbreak,” Professor Sajeev explained. “It takes into account not only the movements of individual animals (as recorded by each animal’s Radio Frequency Identification tag) in transmitting the disease, but also indirect transmission pathways including movements of people, vehicles, and farm produce. In modelling wind-borne spread of the disease we’ve had to allow for the variability of wind speed, temperature and humidity.”
The other members of the UNE team are Dr Paul Kwan, Mitchell Welch, Dr Mark Evered, Professor Graham Leedham, and Dr Ashoka Jayawardena.
The first phase of the project, the prototype of which is now complete, focuses on the spread of the disease up to the point of detection (the “silent spread” phase, during which there are no control measures). The model produces a variety of graphs, tables and maps predicting the number of animals affected, the stages of the disease in each animal, and the pattern of transmission and spread of the disease from day to day.
“The Department contacted our group here at UNE because of our expertise in the fields of biosecurity, software engineering and artificial intelligence, and our involvement in the Australian Biosecurity Intelligence Network,” Professor Sajeev said.
THE PHOTOGRAPH displayed here shows Mitchell Welch, a PhD student in UNE’s School of Science and Technology who, as a Research Assistant on the project, has done most of the programming. It expands to include Professor Sajeev and Dr Paul Kwan.